Pub Date : 2018-11-01DOI: 10.1109/ITSC.2018.8569727
S. Ishida, Jumpei Kajimura, M. Uchino, S. Tagashira, Akira Fukuda
In the ITS (intelligent transportation system), vehicle detection is one of the core technologies. We are developing an acoustic vehicle detector that detects vehicles using a sound map, which is a map of sound arrival time difference on two microphones. We developed vehicle detection algorithms based on state machine and DTW (dynamic time warping) to detect S-curves on a sound map drawn by passing vehicles. However, the detection algorithms often fail to detect simultaneous and sequential passing vehicles. This paper presents SAVeD, a sequential acoustic vehicle detector. The SAVeD fits an S-curve model to sound map points using a RANSAC (random sample consensus) robust estimation method to detect each vehicle. The SAVeD then removes sound map points corresponding to the detected vehicle and continues vehicle detection process for the following vehicles. Experimental evaluations demonstrated that the SAVeD improves detection accuracy by more than 10 points compared to the state-machine based algorithm.
{"title":"SAVeD: Acoustic Vehicle Detector with Speed Estimation capable of Sequential Vehicle Detection","authors":"S. Ishida, Jumpei Kajimura, M. Uchino, S. Tagashira, Akira Fukuda","doi":"10.1109/ITSC.2018.8569727","DOIUrl":"https://doi.org/10.1109/ITSC.2018.8569727","url":null,"abstract":"In the ITS (intelligent transportation system), vehicle detection is one of the core technologies. We are developing an acoustic vehicle detector that detects vehicles using a sound map, which is a map of sound arrival time difference on two microphones. We developed vehicle detection algorithms based on state machine and DTW (dynamic time warping) to detect S-curves on a sound map drawn by passing vehicles. However, the detection algorithms often fail to detect simultaneous and sequential passing vehicles. This paper presents SAVeD, a sequential acoustic vehicle detector. The SAVeD fits an S-curve model to sound map points using a RANSAC (random sample consensus) robust estimation method to detect each vehicle. The SAVeD then removes sound map points corresponding to the detected vehicle and continues vehicle detection process for the following vehicles. Experimental evaluations demonstrated that the SAVeD improves detection accuracy by more than 10 points compared to the state-machine based algorithm.","PeriodicalId":395239,"journal":{"name":"2018 21st International Conference on Intelligent Transportation Systems (ITSC)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121679327","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-11-01DOI: 10.1109/ITSC.2018.8569831
Awad Abdelhalim, M. Abbas
In this study, a cooperative Transit Signal Priority (TSP) strategy utilizing transit vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication is proposed and assessed. The proposed Bus-Holding Transit Signal Priority (BHTSP) is implemented and evaluated for a detailed isolated intersection constructed in VISSIM microsimulation software resembling the intersection of Alumni Mall and Main Street in Blacksburg, VA, USA. The impact of the proposed strategy was assessed for Blacksburg Transit vehicles arriving at the major Squires Eastbound bus stop, using high quality data that includes up-to-date vehicle flows, signal timing, transit schedules, and actual transit arrival, departure and dwell times. An advanced vehicle actuated control logic was implemented using the VISSIM COM Application Programming Interface (API) to emulate communications between transit vehicles and signal controller. When a transit vehicle is approaching the bus stop, the transit vehicle that is dwelling at the stop is forced to hold past its dwell time and wait for the upstream vehicle to arrive and dwell, in order to subsequently generate a priority request that would serve more than one transit vehicle convoying from the stop towards the intersection. This strategy results in an improved TSP performance by reducing the number of priority requests, missed TSP calls, and reducing the adverse effects on non-transit traffic at the main arterial. The proposed BHTSP was further fortified by utilizing V2I communications, the Connected BHTSP (C-BHTSP) strategy has resulted in 61% reduction of transit delay in the network compared to the base scenario, alongside a 38% reduction in early green priority requests, reducing the incurred arterial vehicle delay by 32%, and reducing total additional system stops by 51% compared to conventional TSP strategy.
{"title":"Impact Assessment of a Cooperative Bus-Holding Transit Signal Priority Strategy","authors":"Awad Abdelhalim, M. Abbas","doi":"10.1109/ITSC.2018.8569831","DOIUrl":"https://doi.org/10.1109/ITSC.2018.8569831","url":null,"abstract":"In this study, a cooperative Transit Signal Priority (TSP) strategy utilizing transit vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication is proposed and assessed. The proposed Bus-Holding Transit Signal Priority (BHTSP) is implemented and evaluated for a detailed isolated intersection constructed in VISSIM microsimulation software resembling the intersection of Alumni Mall and Main Street in Blacksburg, VA, USA. The impact of the proposed strategy was assessed for Blacksburg Transit vehicles arriving at the major Squires Eastbound bus stop, using high quality data that includes up-to-date vehicle flows, signal timing, transit schedules, and actual transit arrival, departure and dwell times. An advanced vehicle actuated control logic was implemented using the VISSIM COM Application Programming Interface (API) to emulate communications between transit vehicles and signal controller. When a transit vehicle is approaching the bus stop, the transit vehicle that is dwelling at the stop is forced to hold past its dwell time and wait for the upstream vehicle to arrive and dwell, in order to subsequently generate a priority request that would serve more than one transit vehicle convoying from the stop towards the intersection. This strategy results in an improved TSP performance by reducing the number of priority requests, missed TSP calls, and reducing the adverse effects on non-transit traffic at the main arterial. The proposed BHTSP was further fortified by utilizing V2I communications, the Connected BHTSP (C-BHTSP) strategy has resulted in 61% reduction of transit delay in the network compared to the base scenario, alongside a 38% reduction in early green priority requests, reducing the incurred arterial vehicle delay by 32%, and reducing total additional system stops by 51% compared to conventional TSP strategy.","PeriodicalId":395239,"journal":{"name":"2018 21st International Conference on Intelligent Transportation Systems (ITSC)","volume":"2014 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114719752","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-11-01DOI: 10.1109/ITSC.2018.8569983
Yuxin He, Yang Zhao, Jin Qin, K. Tsui
Transportation network efficiency is a comprehensive reflection of the operation of transportation networks. An effective quantitative evaluation method for the transportation network efficiency is important as it can provide a feedback mechanism of network operation conditions in the process of network design, which gives a theoretical basis for the optimization of urban transportation network. In general, a well-designed transportation network should be adapted to multi-typed traffic demands by considering their characteristics after reconstructing. Thus, on the choice of an effective quantitative evaluation method for the transportation network efficiency, this paper proposes a bi-level programming model with the objective of maximizing transportation network efficiency in mixed network design, which has two lower users' equilibrium models corresponding to two kinds of traffic demands. A hybrid Genetic Algorithm (GA) and Frank-Wolfe Algorithm is then developed to solve the proposed problem. Results of the case study show that the network designed by the proposed model a) results in a more rational distribution of traffic flow, b) improves the adaptability of the transportation network and alleviates the traffic congestion, and c) economizes on the use of land, providing a solid foundation for the sustainable development of transportation network.
{"title":"Efficiency-Based Mixed Network Design Considering Multi-Typed Traffic Demands","authors":"Yuxin He, Yang Zhao, Jin Qin, K. Tsui","doi":"10.1109/ITSC.2018.8569983","DOIUrl":"https://doi.org/10.1109/ITSC.2018.8569983","url":null,"abstract":"Transportation network efficiency is a comprehensive reflection of the operation of transportation networks. An effective quantitative evaluation method for the transportation network efficiency is important as it can provide a feedback mechanism of network operation conditions in the process of network design, which gives a theoretical basis for the optimization of urban transportation network. In general, a well-designed transportation network should be adapted to multi-typed traffic demands by considering their characteristics after reconstructing. Thus, on the choice of an effective quantitative evaluation method for the transportation network efficiency, this paper proposes a bi-level programming model with the objective of maximizing transportation network efficiency in mixed network design, which has two lower users' equilibrium models corresponding to two kinds of traffic demands. A hybrid Genetic Algorithm (GA) and Frank-Wolfe Algorithm is then developed to solve the proposed problem. Results of the case study show that the network designed by the proposed model a) results in a more rational distribution of traffic flow, b) improves the adaptability of the transportation network and alleviates the traffic congestion, and c) economizes on the use of land, providing a solid foundation for the sustainable development of transportation network.","PeriodicalId":395239,"journal":{"name":"2018 21st International Conference on Intelligent Transportation Systems (ITSC)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127687373","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-11-01DOI: 10.1109/ITSC.2018.8569772
Chiyu Dong, J. Dolan
It is essential for autonomous driving cars to understand and predict other surrounding cars' behaviors, especially in urban environments, due to the high traffic volumes and complex interactions. Modeling the interaction among cars and their behaviors is challenging. The behavior estimation of a surrounding car serves as prior knowledge which helps the trajectory planner generate a path to perform properly with the other vehicles. It closes the gap between the high-level decision making and path planning. A new data-driven method is proposed to extend our previous behavior estimation. The new method predicts the continuous lane-change trajectory of a target car by modeling the interaction of all its surrounding vehicles' trajectories, without over-the-air communication between vehicles. The advantages of this approach are: 1. Learning the interactive model from real data; 2. Giving long-horizon estimation of the continuous trajectory of a target vehicle. The method is trained and evaluated on a public dataset. The experimental results show that the proposed method successfully predicts trajectories considering mutual interactions among cars, with low error based on the ground-truth.
{"title":"Continuous Behavioral Prediction in Lane-Change for Autonomous Driving Cars in Dynamic Environments","authors":"Chiyu Dong, J. Dolan","doi":"10.1109/ITSC.2018.8569772","DOIUrl":"https://doi.org/10.1109/ITSC.2018.8569772","url":null,"abstract":"It is essential for autonomous driving cars to understand and predict other surrounding cars' behaviors, especially in urban environments, due to the high traffic volumes and complex interactions. Modeling the interaction among cars and their behaviors is challenging. The behavior estimation of a surrounding car serves as prior knowledge which helps the trajectory planner generate a path to perform properly with the other vehicles. It closes the gap between the high-level decision making and path planning. A new data-driven method is proposed to extend our previous behavior estimation. The new method predicts the continuous lane-change trajectory of a target car by modeling the interaction of all its surrounding vehicles' trajectories, without over-the-air communication between vehicles. The advantages of this approach are: 1. Learning the interactive model from real data; 2. Giving long-horizon estimation of the continuous trajectory of a target vehicle. The method is trained and evaluated on a public dataset. The experimental results show that the proposed method successfully predicts trajectories considering mutual interactions among cars, with low error based on the ground-truth.","PeriodicalId":395239,"journal":{"name":"2018 21st International Conference on Intelligent Transportation Systems (ITSC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126405027","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-11-01DOI: 10.1109/ITSC.2018.8569384
C. Philippe, L. Adouane, B. Thuilot, A. Tsourdos, Hyo-Sang Shin
In this paper is presented a linear MPC controller design for autonomous cars navigation. It combines both the lateral and longitudinal control. The MPC cost function has been designed to account for human driving behaviours, i.e., it smoothes out coarse reference trajectories. Furthermore, a safety monitoring module has been implemented. It computes an estimated time before reaching an unacceptable situation (w.r.t. comfort constraints and tracking performance) under the current tracking conditions. The overall benefit of this controller is to guarantee trajectory smoothness while outputting information on its performance. This information will later be used to re-plan safe trajectories in dynamic environments. The proposed linear MPC controller has been tested in a typical urban scenario based on a realistic simulator.
{"title":"Safe and Online MPC for Managing Safety and Comfort of Autonomous Vehicles in Urban Environment","authors":"C. Philippe, L. Adouane, B. Thuilot, A. Tsourdos, Hyo-Sang Shin","doi":"10.1109/ITSC.2018.8569384","DOIUrl":"https://doi.org/10.1109/ITSC.2018.8569384","url":null,"abstract":"In this paper is presented a linear MPC controller design for autonomous cars navigation. It combines both the lateral and longitudinal control. The MPC cost function has been designed to account for human driving behaviours, i.e., it smoothes out coarse reference trajectories. Furthermore, a safety monitoring module has been implemented. It computes an estimated time before reaching an unacceptable situation (w.r.t. comfort constraints and tracking performance) under the current tracking conditions. The overall benefit of this controller is to guarantee trajectory smoothness while outputting information on its performance. This information will later be used to re-plan safe trajectories in dynamic environments. The proposed linear MPC controller has been tested in a typical urban scenario based on a realistic simulator.","PeriodicalId":395239,"journal":{"name":"2018 21st International Conference on Intelligent Transportation Systems (ITSC)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115917771","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-11-01DOI: 10.1109/ITSC.2018.8569896
Qi Luo, Xuechun Dou, Xuan Di, R. Hampshire
Motivated by the meteoric rise in the adoption of both ride-hailing services (DiDi, Uber, Lyft, etc.) and dockless bikesharing services (Ofo, Mobike, LimeBike, etc.), we propose a multimodal system where passengers ride a dockless bikeshare to/from hubs where they switch modes to/from a carpool. The proposed mutlimodal system is a generalization of the existing Uber ExpressPool service. The goal of this paper is to test empirically the feasibility of the proposed multimodal system. We accomplish this goal with the aid of time-stamped taxi origin/destination data from New York City. The analysis has two steps: network design and trip assignment. First, we identify 17 carpool hub locations with a coverage of 1 km to capture all taxi trip demand within Manhattan during peak hours. After designing the network, we then assign trips to carpools, within each hub, that have similar trip start times and destinations. We formulate the assignment problem as an offline matching algorithm on a bipartite graph. We found that over 80 percent of all trips can be assigned to carpools at almost all hubs. Compared to a single-modal system, the multimodal system served the same number of passengers with 40 percent fewer taxis. We found the matching rate to be consistent for every month in 2015. These results provide initial evidence that multimodal connections between ride-hailing and dockless bikesharing are feasible, reduces passenger trip times, and decreases road congestion.
{"title":"Multimodal Connections between Dockless Bikesharing and Ride-Hailing: An Empirical Study in New York City","authors":"Qi Luo, Xuechun Dou, Xuan Di, R. Hampshire","doi":"10.1109/ITSC.2018.8569896","DOIUrl":"https://doi.org/10.1109/ITSC.2018.8569896","url":null,"abstract":"Motivated by the meteoric rise in the adoption of both ride-hailing services (DiDi, Uber, Lyft, etc.) and dockless bikesharing services (Ofo, Mobike, LimeBike, etc.), we propose a multimodal system where passengers ride a dockless bikeshare to/from hubs where they switch modes to/from a carpool. The proposed mutlimodal system is a generalization of the existing Uber ExpressPool service. The goal of this paper is to test empirically the feasibility of the proposed multimodal system. We accomplish this goal with the aid of time-stamped taxi origin/destination data from New York City. The analysis has two steps: network design and trip assignment. First, we identify 17 carpool hub locations with a coverage of 1 km to capture all taxi trip demand within Manhattan during peak hours. After designing the network, we then assign trips to carpools, within each hub, that have similar trip start times and destinations. We formulate the assignment problem as an offline matching algorithm on a bipartite graph. We found that over 80 percent of all trips can be assigned to carpools at almost all hubs. Compared to a single-modal system, the multimodal system served the same number of passengers with 40 percent fewer taxis. We found the matching rate to be consistent for every month in 2015. These results provide initial evidence that multimodal connections between ride-hailing and dockless bikesharing are feasible, reduces passenger trip times, and decreases road congestion.","PeriodicalId":395239,"journal":{"name":"2018 21st International Conference on Intelligent Transportation Systems (ITSC)","volume":"351 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115955154","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-11-01DOI: 10.1109/ITSC.2018.8569747
A. Alho, Linlin You, Fangping Lu, L. Cheah, Fang Zhao, M. Ben-Akiva
Freight road vehicle operations vary widely depending on a multitude of factors such as industry type, commodities transported or geographical scope. Vehicle tracking is one of the most common approaches to understand operation patterns and it has been facilitated by the increasing availability of GPS-enabled devices. We describe a method that supplements vehicle tracking data with day-to-day driver activity surveys to collect static and dynamic data related to freight vehicle operations. The survey was designed to enable innovative data analysis and modelling. We detail the data collection method demonstrated in Singapore and illustrate three data-driven insights which are of interest in the urban freight domain: (1) freight vehicle overnight parking, (2) tour patterns and associated vehicle usage characteristics, and (3) commodity flow patterns. The unique insights demonstrated by the analyses corroborate the potential of the described data collection method to further understand freight vehicle operations.
{"title":"Next-generation freight vehicle surveys: Supplementing truck GPS tracking with a driver activity survey","authors":"A. Alho, Linlin You, Fangping Lu, L. Cheah, Fang Zhao, M. Ben-Akiva","doi":"10.1109/ITSC.2018.8569747","DOIUrl":"https://doi.org/10.1109/ITSC.2018.8569747","url":null,"abstract":"Freight road vehicle operations vary widely depending on a multitude of factors such as industry type, commodities transported or geographical scope. Vehicle tracking is one of the most common approaches to understand operation patterns and it has been facilitated by the increasing availability of GPS-enabled devices. We describe a method that supplements vehicle tracking data with day-to-day driver activity surveys to collect static and dynamic data related to freight vehicle operations. The survey was designed to enable innovative data analysis and modelling. We detail the data collection method demonstrated in Singapore and illustrate three data-driven insights which are of interest in the urban freight domain: (1) freight vehicle overnight parking, (2) tour patterns and associated vehicle usage characteristics, and (3) commodity flow patterns. The unique insights demonstrated by the analyses corroborate the potential of the described data collection method to further understand freight vehicle operations.","PeriodicalId":395239,"journal":{"name":"2018 21st International Conference on Intelligent Transportation Systems (ITSC)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130184838","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-11-01DOI: 10.1109/ITSC.2018.8569428
Ó. Mata-Carballeira, I. D. Campo, M. V. Martínez, J. Echanobe
This work presents a Deep Extreme Learning Machine with Auto Encoder scheme for Speed Limit Signs Recognition in the field of Advanced Driving Assistance Systems, where traffic sign recognition from video imaging plays an important role specially to provide vehicles with automated speed limits enforcement. Current solutions adopted by car manufacturers do not provide robust enough recognition behaviors when the image quality, the lighting conditions or the clearance of the traffic sign are compromised. These conditions result in misinterpreting of the speed limits, showing wrong on-screen advices which might confuse the driver, causing dangerous situations. In this work, the full chain of operations is studied. The proposed scheme is trained and tested with the German Traffic Sign Recognition Benchmark (GTSRB) database, achieving recognition times as short as 0.62 ms per sample, reaching with this timing real-time operation, and an accuracy of up to 92% with a simpler structure than other techniques currently used, such as Convolutional Neural Networks (CNNs).
{"title":"Deep Extreme Learning Machines with Auto Encoder for Speed Limit Signs Recognition","authors":"Ó. Mata-Carballeira, I. D. Campo, M. V. Martínez, J. Echanobe","doi":"10.1109/ITSC.2018.8569428","DOIUrl":"https://doi.org/10.1109/ITSC.2018.8569428","url":null,"abstract":"This work presents a Deep Extreme Learning Machine with Auto Encoder scheme for Speed Limit Signs Recognition in the field of Advanced Driving Assistance Systems, where traffic sign recognition from video imaging plays an important role specially to provide vehicles with automated speed limits enforcement. Current solutions adopted by car manufacturers do not provide robust enough recognition behaviors when the image quality, the lighting conditions or the clearance of the traffic sign are compromised. These conditions result in misinterpreting of the speed limits, showing wrong on-screen advices which might confuse the driver, causing dangerous situations. In this work, the full chain of operations is studied. The proposed scheme is trained and tested with the German Traffic Sign Recognition Benchmark (GTSRB) database, achieving recognition times as short as 0.62 ms per sample, reaching with this timing real-time operation, and an accuracy of up to 92% with a simpler structure than other techniques currently used, such as Convolutional Neural Networks (CNNs).","PeriodicalId":395239,"journal":{"name":"2018 21st International Conference on Intelligent Transportation Systems (ITSC)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134527494","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-11-01DOI: 10.1109/ITSC.2018.8569655
Julian Schwehr, Volker Willert
For a safe handover of the driving task or driver-adaptive warning strategies the driver's situation awareness is a helpful source of information. In order to estimate and track the driver's focus of attention over time in a dynamic automotive scene, a Multi-Hypothesis Multi-Model probabilistic tracking framework was developed in which we postulate consistency between machine and human perception during gaze fixations. Within this framework, we explicitly included target object motion in the spatial transition step and integrated spatiotemporal models of human-like gaze behavior for fixations and saccades in the motion transition. This elaborate design makes the target estimation robust and yet flexible. At the same time, the representation in continuous 2D coordinates makes the algorithm run in real time on a standard laptop. By incorporating dynamic and static potential gaze targets from an object list and a free space spline, the algorithm is in principle independent from the applied sensor setup. The benefit of the proposed model is presented on real world data where the filter's tracking performance as well as the driver's visual sampling are presented based on an exemplary scene.
{"title":"Multi-Hypothesis Multi-Model Driver's Gaze Target Tracking","authors":"Julian Schwehr, Volker Willert","doi":"10.1109/ITSC.2018.8569655","DOIUrl":"https://doi.org/10.1109/ITSC.2018.8569655","url":null,"abstract":"For a safe handover of the driving task or driver-adaptive warning strategies the driver's situation awareness is a helpful source of information. In order to estimate and track the driver's focus of attention over time in a dynamic automotive scene, a Multi-Hypothesis Multi-Model probabilistic tracking framework was developed in which we postulate consistency between machine and human perception during gaze fixations. Within this framework, we explicitly included target object motion in the spatial transition step and integrated spatiotemporal models of human-like gaze behavior for fixations and saccades in the motion transition. This elaborate design makes the target estimation robust and yet flexible. At the same time, the representation in continuous 2D coordinates makes the algorithm run in real time on a standard laptop. By incorporating dynamic and static potential gaze targets from an object list and a free space spline, the algorithm is in principle independent from the applied sensor setup. The benefit of the proposed model is presented on real world data where the filter's tracking performance as well as the driver's visual sampling are presented based on an exemplary scene.","PeriodicalId":395239,"journal":{"name":"2018 21st International Conference on Intelligent Transportation Systems (ITSC)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134331314","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-11-01DOI: 10.1109/ITSC.2018.8569290
M. Aljamal, H. Rakha, Jianhe Du, Ihab El-Shawarby
Evacuation activities can be evaluated using different simulation models. However, recently, microscopic simulation models have become a more popular tool for this purpose. The objectives of this study are to model multiple evacuation scenarios and to compare a microscopic traffic simulation tool (in this case INTEGRATION) against a mesoscopic traffic simulation tool (MATSim). Given that the demand was the same for both models, the comparison was achieved based on two indicators: estimated evacuation time and average trip duration. The results show that the estimated evacuation times in both models are similar since the input traffic demand governed this measure. However, the evaluation also shows a considerable difference between the two models in the average trip duration. The microscopic traffic simulation tool produces logical results with trip durations increasing with increased traffic demand levels and decreasing road capacity scenarios, whereas the average trip duration using the mesoscopic simulation tool decreases with increasing demand levels and increasing road capacity scenarios.
{"title":"Comparison of Microscopic and Mesoscopic Traffic Modeling Tools for Evacuation Analysis","authors":"M. Aljamal, H. Rakha, Jianhe Du, Ihab El-Shawarby","doi":"10.1109/ITSC.2018.8569290","DOIUrl":"https://doi.org/10.1109/ITSC.2018.8569290","url":null,"abstract":"Evacuation activities can be evaluated using different simulation models. However, recently, microscopic simulation models have become a more popular tool for this purpose. The objectives of this study are to model multiple evacuation scenarios and to compare a microscopic traffic simulation tool (in this case INTEGRATION) against a mesoscopic traffic simulation tool (MATSim). Given that the demand was the same for both models, the comparison was achieved based on two indicators: estimated evacuation time and average trip duration. The results show that the estimated evacuation times in both models are similar since the input traffic demand governed this measure. However, the evaluation also shows a considerable difference between the two models in the average trip duration. The microscopic traffic simulation tool produces logical results with trip durations increasing with increased traffic demand levels and decreasing road capacity scenarios, whereas the average trip duration using the mesoscopic simulation tool decreases with increasing demand levels and increasing road capacity scenarios.","PeriodicalId":395239,"journal":{"name":"2018 21st International Conference on Intelligent Transportation Systems (ITSC)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131519224","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：